The HIV epidemic in South Africa has several distinctive features; "explosive" spread, predominance in women at younger ages, and very high prevalence with no sign of a 'saturation' plateau. As accompanying morbidity and now mortality rise, design of effective prevention and treatment calls for better understanding of the unfolding epidemiologic pattern. Community-level epidemiological, demographic and social data are necessary as the basis for assessing prevention and/or therapeutic approaches. We therefore propose to monitor epidemic trends in a rural community by gender, age and relevant social factors such as occupation and education at four levels: population rates of HIV prevalence and incidence; morbidity and health service impact; mortality; and the social/behavioral impact of AIDS mortality. HIV prevalence will be measured in patients attending antenatal, family planning, sexually transmitted diseases and tuberculosis clinics at two major health centers in a rural South Africa district. Each of these risk groups will be sampled for two months each year in order to obtain adequate numbers of or age-specific rates. The sensitive/less sensitive assay, which was recently validated with South African seroconvertors, will be used to determine the incidence rates in these groups. HIV-associated morbidity will be documented in clinic registers specifically devised for this purpose. These data will then elucidate the case mix at the clinics and provide an indication of the changing burden on AIDS in primary health care services. Data on deaths due to AIDS will be collected from several sources, including death certificates, cemetery records, undertaker records and from local traditional leaders. Trends in mortality and the changing age-specific mortality rates will be documented together with more detailed information obtained from verbal autopsies. Household members experiencing a death due to AIDS will be compared to household members experiencing a death due to any other cause, to determine if the AIDS death led to the adoption of HIV riskreducing behavior. Taken together these data will produce a comprehensive profile of the evolving HIV/AIDS epidemic in a rural community in South Africa.